1. Field
This disclosure relates to wireless communication networks.
2. Description of the Related Art
It is well known that a significant power saving can be achieved in a sensor network by having intermediate sensors act as routing points for messages when transferring data. Since sensor networks are controlled by a central authority with an objective of minimizing power consumption, that authority can ensure that intermediate sensors consume their energy for the good of the network. The primary advantage of having intermediate nodes forward messages is that the power consumption is a non-linear function of distance (depending on the network and type of communication it is assumed to be between squared and quadratic). Thus a series of communications over short distances or short hops requires less power than a single communication over the total distance.
These laws of physics govern other networks including cellular phones communicating with communication towers and peer-to-peer (P2P) networking via WiFi for laptops. However, the nodes on these networks are controlled by individuals, rather than by a central authority.
Within this description, the term “mesh network” will mean a communication network in which at least some nodes are capable of forwarding messages for other nodes. The determination of which nodes in a mesh network forward messages may be made dynamically depending on the location of the nodes and the network connectivity.
Within this description, the term “selfish user” will be used to describe a network node controlled by a party who wishes to optimize their own situation, rather than the well-being of the network. The term “altruistic user” will be used to describe a network node that will always forward messages. Mesh networks made up of selfish users, at least in part, will be called “selfish networks”. A significant power consumption savings may also be achieved by using short hops to route messages among participants in selfish wireless networks such as cellular phone networks and WiFi networks.
There are also many other advantages to having users forward messages in a mesh network. Forwarding may allow messages to reach places that might not be served by central towers (or that can be designed not to be served by central towers based on forwarding). One example is the One Laptop Per Child project, which intends to extend coverage to a central server via forwarding. Another advantage of message forwarding is that it allows for a better reuse of frequency to serve more users with a smaller amount of bandwidth.
However, there are several practical objections to use of a mesh network when there are selfish users. The first objection concerns the perception that routing might compromise privacy. Since users feel that the forwarding entity has some special access to their data, they naturally feel that their privacy is threatened. The fear of privacy loss can be dealt with through the use of cryptography. By using the encryption modules in cellular phones or by adding encryption into peripheral devices (such as headphones) we can ensure that the problem of privacy does not appear as an ongoing threat to users.
Another problem that might arise when using forwarding through nodes controlled by individuals is the issue of reliability. This can be dealt with by demanding acknowledgments (possibly with a signature) to ensure that each message was received by its intended destination.
The main obstacle to the deployment of message forwarding protocols in wireless networks controlled by individuals is the reluctance of selfish users to give up some of their own power storage in order to contribute to the good of the network. A network that forwards messages incurs an “energy cost” for intermediate nodes in the message route. In effect, when one is not talking, one's battery is used for transporting messages for other network users. These energy costs might incentivize the owners of required intermediate nodes to refuse to forward messages, hence making it impossible to achieve an overall system-wide energy saving.
Prior work has concentrated on utilizing a system of micro-payments in order to motivate selfish users to participate in message forwarding on a network. Micro-payments are complicated for users to understand. Furthermore, existing schemes for micro-payments may require users to constantly adjust their willingness to forward based on constraints, such as when users intend to recharge their phones and how much talking they intend to do.